Inputs & Outputs (split from BGOL thread)

[fungus]

No thanks, that was fine. It is, in effect, block changes then. I've been doing those since my 3.8 Jaguar days, commencing in 1968: start in 1st., early change to 2nd., bit of boot, then change to 4th., bit more boot, then engage the overdrive. Happy days.

So you were advanced, and years ahead of your time then.

[waremark]

A diesel I have will pull like a train in 6th from about 30mph.
There would be no point whatsoever in normal driving in using the gears in any set order; after all they give you 6 to chose from.
The correct one that will complete the job is what you are after whilst keeping the drive train in pretty much a straight line

Drive train in a straight line?? Don't understand that last line.

[Silk]

In other words, you look to see if it's pretty and don't care if you end up embedded in a tree. I'd consider that a total abrogation of responsibility.

Your words, not mine. If you end up "embedded in a tree", then it's not a good input. Good inputs certainly can look pretty, as can good outputs, but that's not what makes them good.

Your terminology is somewhat confused here and is making the point easy to miss. As far as I can see your definition of a good input for steering, say, is pull-push to the right steering angle for the bend, but you have no method of teaching what the right angle for the bend is, how to be able to detect that, what feedback to try and receive from the vehicle.

I haven't mentioned a particular steering method - that's already been done to death elsewhere. I'm simply challenging the position that it's only the outputs that matter. To me, that's fundamentally flawed for a number of reasons that I've already given. If it's obvious that an output isn't working quite right, then it's logical to look for a problem with the input. It may be down to the method or the execution of a particular method. It makes no sense at all to say it doesn't matter.

There's a whole panoply of things you're either missing or leaving up to your tutee to get right on their own because you're hooked up on how they should be operating the controls.

I'm not aware that I'm "hooked up" on anything. In fact, I'm not even arguing that it's ONLY the inputs that matter. It's you arguing that it's ONLY the outputs.

[Gareth]

I'm simply challenging the position that it's only the outputs that matter. To me, that's fundamentally flawed for a number of reasons that I've already given. If it's obvious that an output isn't working quite right, then it's logical to look for a problem with the input.

Good point, but the question is when you need to consider the inputs. Do you need to consider the inputs if the outputs are good? If so, why?

[Horse]

I'm simply challenging the position that it's only the outputs that matter. To me, that's fundamentally flawed for a number of reasons that I've already given. If it's obvious that an output isn't working quite right, then it's logical to look for a problem with the input.

Good point, but the question is when you need to consider the inputs. Do you need to consider the inputs if the outputs are good? If so, why?

And, given the motorcycling example I gave earlier, how do you actually assess the outputs?

I'll give you a specific motorcycling example, where there needs to be a specific test requirement for a particular motorcycle control skill:

In the old 'Blue Book' m/c Roadcraft, there was no mention of how to steer a motorcycle, from memory the wording was along the lines of 'simply lean'. In actual fact, that is tosh, a bike's (whether pedal or motor) turn is initiated by counter-steering - a brief and slight initial turn of the handlebars (and so front wheel) in the opposite direction to that which you wish to go.

By and large, riders got around corners by leaning. Huh? Haven't I just said they couldn't? What they actually did was to move their upper body to one side (to try and 'lean' the bike) - but accidentally pressed forwards on that end of the bar . . . so initiating the turn.

The problem comes when you wish to make a very immediate turn, such as has been introduced in the last couple of years to UK bike testing, with a 'swerve' manoeuvre. Cue riders now being taught how to steer . . . A classic example of where police influence at the higher echelons of UK training has held us back compared to much of the rest of the world - but, worse still, potentially to the lives of many motorcyclists who weren't taught how to steer a bike (and so may have run wide in corners etc.) because "it isn't in Roadcraft".

Now, returning to input - output.

You can steer 'accidentally', and [usually] achieve a turn.

Or:
You could 'pull back' on the 'outside' bar end.
You could press 'forwards' on the 'inside' bar end.

But how will I, as an assessor following that rider, know?

[sussex2]

A diesel I have will pull like a train in 6th from about 30mph.
There would be no point whatsoever in normal driving in using the gears in any set order; after all they give you 6 to chose from.
The correct one that will complete the job is what you are after whilst keeping the drive train in pretty much a straight line

Drive train in a straight line?? Don't understand that last line.

It is to engage a gear with the engine under power - matched revs which put less strain on the mechanics and waste less energy. It is not always possible but generally preferable.

[zadocbrown]

I think if I were starting from the current position of the IAM, I'd be doing the same and risk sacrificing the top 5-10% of Observers for the sake of consistency and quality control.

I guess if you were a banana farmer you'd throw away all the bent ones then?

I'd rather get rid of the 90/95% and start again with the 5/10% who know what the hell they're doing.....

[hir]

I haven't mentioned a particular steering method - that's already been done to death elsewhere. I'm simply challenging the position that it's only the outputs that matter. To me, that's fundamentally flawed for a number of reasons that I've already given. If it's obvious that an output isn't working quite right, then it's logical to look for a problem with the input. It may be down to the method or the execution of a particular method. It makes no sense at all to say it doesn't matter.

I think there is a bit of confusion here which is obscuring the nature of the inputs/outcomes debate. Nobody is arguing that inputs and their skilful application do not matter. As you say... "If it's obvious that an output isn't working quite right, then it's logical to look for a problem with the input." The argument is about the mandatory use, or non-use, of prescribed inputs in order to achieve a satisfactory [read good] outcome.

By assessing outcomes it leaves one free to use those inputs that suit one best. There may be many different inputs that might achieve a good outcome. Those who advocate judging inputs, in my experience, advocate just one allowable type of input. Any other method of input, however good the outcome, is anathema. That can't be right!

[TripleS]

No thanks, that was fine. It is, in effect, block changes then. I've been doing those since my 3.8 Jaguar days, commencing in 1968: start in 1st., early change to 2nd., bit of boot, then change to 4th., bit more boot, then engage the overdrive. Happy days.

So you were advanced, and years ahead of your time then.

Oh hell, I don't know about that; but I'm now rather more concerned about slipping backwards.

[Silk]

No thanks, that was fine. It is, in effect, block changes then. I've been doing those since my 3.8 Jaguar days, commencing in 1968: start in 1st., early change to 2nd., bit of boot, then change to 4th., bit more boot, then engage the overdrive. Happy days.

So you were advanced, and years ahead of your time then.

Oh hell, I don't know about that; but I'm now rather more concerned about slipping backwards.

Might I suggest the handbrake?

[TripleS]

I haven't mentioned a particular steering method - that's already been done to death elsewhere. I'm simply challenging the position that it's only the outputs that matter. To me, that's fundamentally flawed for a number of reasons that I've already given. If it's obvious that an output isn't working quite right, then it's logical to look for a problem with the input. It may be down to the method or the execution of a particular method. It makes no sense at all to say it doesn't matter.

I think there is a bit of confusion here which is obscuring the nature of the inputs/outcomes debate. Nobody is arguing that inputs and their skilful application do not matter. As you say... "If it's obvious that an output isn't working quite right, then it's logical to look for a problem with the input." The argument is about the mandatory use, or non-use, of prescribed inputs in order to achieve a satisfactory [read good] outcome.

By assessing outcomes it leaves one free to use those inputs that suit one best. There may be many different inputs that might achieve a good outcome. Those who advocate judging inputs, in my experience, advocate just one allowable type of input. Any other method of input, however good the outcome, is anathema. That can't be right!

Surely, to make the best sense of this, we need to monitor things over a fairly lengthy period.

If, over this fairly lengthy period, the driver being monitored is able to produce consistently good outputs, it would suggest to me that he is applying pretty good inputs. If, on the other hand , the outputs are variable (and sometimes poor), then we should look at the inputs and see where the weaknesses lie.

[waremark]

I'm not sure there is any confusion. To put it simply: we're looking at what the driver is doing to the controls versus what the car is doing on the road.

In other words, you look to see if it's pretty and don't care if you end up embedded in a tree. I'd consider that a total abrogation of responsibility.

'If it's pretty' reminds me of an instructor I had in Sweden on an ice driving course. Knowing we were 'advanced drivers' from the UK, he and his colleagues told us forcefully not to use the steering technique they expected to see - PP, which they called 'milking the cow'. I thought this guy did not understand what PP actually was, and said 'let me show you how PP is done properly''. 'Very pretty', he said, 'but what's the point'?'

He had sublime car control. He was pretty good at steering with one hand - or even no hands!

But of course this is totally irrelevant to the thread.

[waremark]

A diesel I have will pull like a train in 6th from about 30mph.
There would be no point whatsoever in normal driving in using the gears in any set order; after all they give you 6 to chose from.
The correct one that will complete the job is what you are after whilst keeping the drive train in pretty much a straight line

Drive train in a straight line?? Don't understand that last line.

It is to engage a gear with the engine under power - matched revs which put less strain on the mechanics and waste less energy. It is not always possible but generally preferable.

So most of us would use the terminology 'Rev-matched block gear changes'. And most of us use them, changing gear directly from wherever we are in the gearbox to whichever gear has now become most appropriat - in some cases 1st to 6th, as I think you suggest.

[TripleS]

I'm not sure there is any confusion. To put it simply: we're looking at what the driver is doing to the controls versus what the car is doing on the road.

In other words, you look to see if it's pretty and don't care if you end up embedded in a tree. I'd consider that a total abrogation of responsibility.

'If it's pretty' reminds me of an instructor I had in Sweden on an ice driving course. Knowing we were 'advanced drivers' from the UK, he and his colleagues told us forcefully not to use the steering technique they expected to see - PP, which they called 'milking the cow'. I thought this guy did not understand what PP actually was, and said 'let me show you how PP is done properly''. 'Very pretty', he said, 'but what's the point'?'

He had sublime car control. He was pretty good at steering with one hand - or even no hands!

But of course this is totally irrelevant to the thread.

....but interesting, and a touch amusing, as well.

[TripleS]

Might I suggest the handbrake?

You might; but that would be focussing on just one input, whereas I'm more concerned about an overall deterioration of a wide range of outputs.